Neural networks for the prediction of spirometric reference values

Taxiarchis Botsis, S. Halkiotis

Research output: Contribution to journalArticle

Abstract

Normal lung function values are conventionally calculated according to prediction equations. The primary objective of this study is the development of a different method for the prediction of FVC and FEV 1 parameters, in order to achieve better correlation of the predicted values to the real ones. Using a sample from the Greek elderly population that was separated into two groups (a training and a testing one), a number of artificial neural networks were trained. Considering that men and women were studied separately and that two parameters (FVC, FEV 1) were the target of the study, four cases came up. In each case two neural networks were trained using different transfer functions, number of neurons and number of layers. When passing the inputs of the testing data set to the trained networks it was found that the outputs were well correlated with the corresponding measures of the sample. Furthermore, the match with the sample, for a number of neural networks developed, was better compared to the matches of Baltopoulos et al. study that used the same sample for developing prediction equations. This high match allows the potential use of neural networks for predicting not only FVC and FEV 1 but also other spirometric parameters.

Original languageEnglish (US)
Pages (from-to)299-309
Number of pages11
JournalInformatics for Health and Social Care
Volume28
Issue number4
DOIs
StatePublished - Dec 4 2003
Externally publishedYes

Fingerprint

Reference Values
Neurons
Lung
Population
Datasets

Keywords

  • Neural networks
  • Prediction
  • Reference values
  • Spirometry

ASJC Scopus subject areas

  • Health Informatics
  • Nursing (miscellaneous)
  • Health Information Management

Cite this

Neural networks for the prediction of spirometric reference values. / Botsis, Taxiarchis; Halkiotis, S.

In: Informatics for Health and Social Care, Vol. 28, No. 4, 04.12.2003, p. 299-309.

Research output: Contribution to journalArticle

@article{e458b3d2cadb4b64bb06a6b8eab0760d,
title = "Neural networks for the prediction of spirometric reference values",
abstract = "Normal lung function values are conventionally calculated according to prediction equations. The primary objective of this study is the development of a different method for the prediction of FVC and FEV 1 parameters, in order to achieve better correlation of the predicted values to the real ones. Using a sample from the Greek elderly population that was separated into two groups (a training and a testing one), a number of artificial neural networks were trained. Considering that men and women were studied separately and that two parameters (FVC, FEV 1) were the target of the study, four cases came up. In each case two neural networks were trained using different transfer functions, number of neurons and number of layers. When passing the inputs of the testing data set to the trained networks it was found that the outputs were well correlated with the corresponding measures of the sample. Furthermore, the match with the sample, for a number of neural networks developed, was better compared to the matches of Baltopoulos et al. study that used the same sample for developing prediction equations. This high match allows the potential use of neural networks for predicting not only FVC and FEV 1 but also other spirometric parameters.",
keywords = "Neural networks, Prediction, Reference values, Spirometry",
author = "Taxiarchis Botsis and S. Halkiotis",
year = "2003",
month = "12",
day = "4",
doi = "10.1080/14639230310001621701",
language = "English (US)",
volume = "28",
pages = "299--309",
journal = "Informatics for Health and Social Care",
issn = "1753-8157",
publisher = "Informa Healthcare",
number = "4",

}

TY - JOUR

T1 - Neural networks for the prediction of spirometric reference values

AU - Botsis, Taxiarchis

AU - Halkiotis, S.

PY - 2003/12/4

Y1 - 2003/12/4

N2 - Normal lung function values are conventionally calculated according to prediction equations. The primary objective of this study is the development of a different method for the prediction of FVC and FEV 1 parameters, in order to achieve better correlation of the predicted values to the real ones. Using a sample from the Greek elderly population that was separated into two groups (a training and a testing one), a number of artificial neural networks were trained. Considering that men and women were studied separately and that two parameters (FVC, FEV 1) were the target of the study, four cases came up. In each case two neural networks were trained using different transfer functions, number of neurons and number of layers. When passing the inputs of the testing data set to the trained networks it was found that the outputs were well correlated with the corresponding measures of the sample. Furthermore, the match with the sample, for a number of neural networks developed, was better compared to the matches of Baltopoulos et al. study that used the same sample for developing prediction equations. This high match allows the potential use of neural networks for predicting not only FVC and FEV 1 but also other spirometric parameters.

AB - Normal lung function values are conventionally calculated according to prediction equations. The primary objective of this study is the development of a different method for the prediction of FVC and FEV 1 parameters, in order to achieve better correlation of the predicted values to the real ones. Using a sample from the Greek elderly population that was separated into two groups (a training and a testing one), a number of artificial neural networks were trained. Considering that men and women were studied separately and that two parameters (FVC, FEV 1) were the target of the study, four cases came up. In each case two neural networks were trained using different transfer functions, number of neurons and number of layers. When passing the inputs of the testing data set to the trained networks it was found that the outputs were well correlated with the corresponding measures of the sample. Furthermore, the match with the sample, for a number of neural networks developed, was better compared to the matches of Baltopoulos et al. study that used the same sample for developing prediction equations. This high match allows the potential use of neural networks for predicting not only FVC and FEV 1 but also other spirometric parameters.

KW - Neural networks

KW - Prediction

KW - Reference values

KW - Spirometry

UR - http://www.scopus.com/inward/record.url?scp=84860394685&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84860394685&partnerID=8YFLogxK

U2 - 10.1080/14639230310001621701

DO - 10.1080/14639230310001621701

M3 - Article

VL - 28

SP - 299

EP - 309

JO - Informatics for Health and Social Care

JF - Informatics for Health and Social Care

SN - 1753-8157

IS - 4

ER -